STS-Lora-Fine-Tuning-Capstone-roberta-base-filtered-137-with-higher-r-mid

This model is a fine-tuned version of FacebookAI/roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7478
• Accuracy: 0.6891

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 3e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	113	1.0392	0.4551
No log	2.0	226	1.0228	0.4944
No log	3.0	339	0.9681	0.5225
No log	4.0	452	0.8498	0.6404
0.8868	5.0	565	0.7953	0.6704
0.8868	6.0	678	0.7626	0.6723
0.8868	7.0	791	0.7913	0.6704
0.8868	8.0	904	0.7607	0.6723
0.6501	9.0	1017	0.7982	0.6873
0.6501	10.0	1130	0.7419	0.6685
0.6501	11.0	1243	0.7451	0.6873
0.6501	12.0	1356	0.7471	0.6760
0.6501	13.0	1469	0.7549	0.6816
0.5977	14.0	1582	0.7364	0.6835
0.5977	15.0	1695	0.7431	0.6760
0.5977	16.0	1808	0.7545	0.6742
0.5977	17.0	1921	0.7556	0.6873
0.5673	18.0	2034	0.7427	0.6873
0.5673	19.0	2147	0.7442	0.6873
0.5673	20.0	2260	0.7600	0.6798
0.5673	21.0	2373	0.7381	0.6854
0.5673	22.0	2486	0.7480	0.6873
0.5561	23.0	2599	0.7489	0.6854
0.5561	24.0	2712	0.7481	0.6873
0.5561	25.0	2825	0.7470	0.6873
0.5561	26.0	2938	0.7530	0.6891
0.5381	27.0	3051	0.7455	0.6854
0.5381	28.0	3164	0.7478	0.6891
0.5381	29.0	3277	0.7483	0.6891
0.5381	30.0	3390	0.7478	0.6891

Framework versions

• PEFT 0.10.0
• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2